Exploiting neighboring-group interactions for the self-selection of a catalytic unit

Article abstract of DOI:10.1002/anie.200703857

(Figure Presented) A good neighbor is better than a far-away friend: A tethering strategy is used to self-select groups that assist in the cleavage of a neighboring carboxylic ester moiety (see picture, TSA=transition-state analogue). A correlation is observed between the amplification at thermodynamic equilibrium and the catalytic efficiency.

Full text of DOI:10.1002/anie.200703857

Angewandte
Chemie
DOI: 10.1002/anie.200703857
Dynamic Selection
Exploiting Neighboring-Group Interactions for the Self-Selection of a
Catalytic Unit**
Giulio Gasparini, Leonard J. Prins,* and Paolo Scrimin*
Dynamic combinatorial chemistry (DCC) is based on the
principle that the thermodynamic composition of a dynamic
library of molecules, that is, a library of which the components
are held together either by noncovalent bonds or reversible
covalent bonds, spontaneously changes upon the input of an
external stimulus.[1] This can be either the addition of a target
molecule, but also an alteration of the environment (pH, light,
etc.). Ideally, the composition of the library changes in favor
of the component that is the most stable under the changed
conditions.[2] In the past decade, DCC has emerged as a
powerful tool for the discovery of, sometimes very surprising,
molecular receptors and novel materials.[3]
In principle, DCC could be applied to the selection of a
catalyst by shifting the equilibrium of the library with
amplification of a molecular receptor for a transition state
of a given reaction.[4] By decreasing the energy of the
transition state by formation of a complex with this molecular
receptor (that is, a catalyst), the reaction rate is obviously
accelerated. This concept was first developed by Pauling,[5]
and applied to catalyst discovery with catalytic antibodies[6]
and imprinted polymers.[7] As a transition state is an elusive
species, a stable analogue is required possessing similar
features in terms of shape and charge distribution. However,
despite the success of DCC, its use for catalyst discovery is
significantly lagging, as evidenced by a very limited number of
publications and, generally, very modest rate accelerations.[4]
This fact suggests that the endeavor is very challenging. In
analogy with enzyme catalysis, an ideal catalyst should first
bind to the substrate and subsequently transform it to
product.[8] Accordingly, the catalyst should both recognize
the substrate and the transition state, although the thermo-
dynamic stabilization of the latter must be much higher. It is
not surprising that in enzymes the substrate and transition
state recognition loci are quite often different because of the
different tasks they have to accomplish.[9] Herein we present
the dynamic self-selection of a functional group which induces
a 60-fold rate enhancement in the basic hydrolysis of a
neighboring carboxylic ester; that is, the selection of a
catalytic unit on the way to the selection of a fully-fledged
catalyst.
Recently, the “tethering” strategy has emerged as a
powerful tool for the detection of weak, noncovalent inter-
actions between substrates and a target.[10] This approach
implies that the target molecule is covalently linked to a
scaffold molecule which has the additional ability to interact
in a reversible manner with library members (Scheme 1). In
this way, the recognition event between target and library
component becomes intramolecular, which, for entropic
reasons, significantly enhances the sensitivity of the screening
process. In fact, Houk has recently pointed out that among the
most efficient enzymes are those that covalently bind the
substrate before its transformation into products.[11]
During studies on hydrazone-based libraries, we recently
observed that the presence of a phosphonate group in 1
resulted in the preferential incorporation of hydrazide B with
respect to A, owing to an intramolecular, electrostatic
interaction between the oppositely charged phosphonate
and ammonium groups.[12] The phosphonate group was
chosen as a target because it is a model for the transition
state of a carboxylic ester hydrolysis. Following the above
concept that stabilization of the transition state should lead to
an increased rate of hydrolysis, we argued that the phospho-
nate group in 1 could be used to self-select hydrazides that
would enhance the cleavage rates of the corresponding
carboxylic ester. Thus, we have screened a library of nine
hydrazides, and present herein compelling data showing the
existence of a correlation between thermodynamic amplifi-
cation in the dynamic screening and the efficiency in assisting
in intramolecular catalysis.
The nine components of the library were chosen from
commercially available hydrazides, of which eight could
potentially interact with a phosphonate moiety, either by
electrostatic interactions (B, C) or the formation of one or
more hydrogen-bonds (D–I) (Scheme 1). Hydrazide A was
not expected to interact with the target and was used as an
internal standard. We also screened aldehyde 2, which
contains a neutral methoxy group: the resulting library
served as a neutral reference to determine the intrinsic
stabilities of the hydrazones in the absence of the target. Any
shift in the library composition using scaffold 1 with respect to
that obtained using scaffold 2 can then be ascribed to an
intramolecular stabilization between the hydrazide and the
phosphonate target.[13]
[*] G. Gasparini, Dr. L. J. Prins, Prof. Dr. P. Scrimin
Department of Chemical Sciences, University of Padova
and CNR ITM, Padova Section
Via Marzolo 1, 35131 Padova (Italy)
Fax: (+39)049-827-5239
E-mail: leonard.prins@unipd.it
Library equilibration studies were performed by adding
either scaffold 1 or 2 (5 mm) to a mixture of hydrazides A–I
(each 1.5 equivalent) in CD3OD. The mixtures were kept at
508C until the thermodynamic equilibrium was reached,
which was detected by the absence of any further change in
paolo.scrimin@unipd.it
[**] We acknowledge financial support from the University of Padova
(CPDA054893) and MIUR (PRIN2006).
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
Angew. Chem. Int. Ed. 2008, 47, 2475 –2479
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2475

Communications
reference hydrazide A (Scheme 2). In this
case, the competition experiments were
performed by adding scaffold
1 or 2
(5 mm) to a solution containing hydrazide
A (5 equiv) and one of the hydrazides B–I
(5 equiv) in CD3OD. The obtained amplifi-
cation factors confirm the trend observed
for the full library screening (Figure 1, dark
gray bars).[16] Furthermore, the data now
substantiate the previously uncertain am-
plification of hydrazone 1I. Considering the
fact that the screening was performed under
conditions where amplification effects are
not maximal,[17] we could confirm that three
of the eight hydrazides (B, C, and I) were
preferentially selected, and B (1.8) to a
higher extent with respect to C and I (1.3
each).
These results clearly demonstrate that
hydrazones 1B, 1C, and 1I are stabilized as
a consequence of an intramolecular inter-
action between the phosphonate and the
functional group present in the hydrazide
unit. This is supported by the fact that the
addition of a phosphate to a hydrazone
Scheme 1. Scaffolds 1, 2, reference hydrazide A, and the hydrazide library (B–I). TBA=tetra-
n-butylammonium.
1
the H NMR spectra of the mixtures (typically 12 hours for
scaffold 1, and 3 days for scaffold 2). All hydrazones are
characterized by the presence of a signal in the d = 8–9 ppm
fingerprint region of the 1H NMR spectrum originating from
À
the hydrazone C H proton. A direct determination of the
library distribution by integration of these signals was not
possible owing to a partial overlap and the presence of signals
in this region from free hydrazide C. This problem could be
1
resolved by measuring H–13C HSQC-spectra of both mix-
tures. The additional separation of signals in the 13C dimen-
sion allowed the individuation and quantification of each
hydrazone present.[14] Concentrations of 1B–I and 2B–I were
determined relative to 1A and 2A, respectively, after which
the amplification was calculated (Figure 1, light gray bars).[15]
The data reveal that charged hydrazones 1B and 1C are
amplified in the mixture.
Figure 1. Observed amplification for each hydrazide in reference to
hydrazide A obtained either from 1H–13C HSQC-spectra (light gray
bars) or from separate mixing experiments (dark gray bars).
library obtained from hydrazides A and B and benzaldehyde
To confirm the observed amplification factors and to did not result in any detectable change of the composition at
maximize precision in this proof-of-concept study, all the
hydrazones were also individually screened against the
thermodynamic equilibrium (data not shown). Following our
hypothesis, the positioning of these functional groups near an
Scheme 2. Competition experiments. All experiments were performed using scaffold 1 (or 2, 5 mm), hydrazide A and either one of the hydrazides
B–I in a 1:5:5 ratio in CD3OD at 508C.
2476 www.angewandte.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 2475 –2479

Angewandte
Chemie
ester moiety should consequently result in an enhanced
cleavage rate of this ester because of transition state
stabilization. In addition, the extent of such a catalytic
effect should be correlated to the extent of amplification
observed (B > C ꢀ I > reference A). To establish such a
correlation, we studied hydrazides B and I in detail, as they
should express a different type and strength of interaction
with the transition state. Therefore, compounds 3B, 3I and
3A (which serves as a reference) were prepared in which the
structural elements of hydrazides A, B, and I, were positioned
in close proximity to a neighboring carboxylate ester
(Scheme 3). Compared to the parent hydrazone structures,
Figure 2. Changes of the absorbance at 280 nm upon a) methanolysis
and b) hydrolysis of compounds 3A, 3B, and 3I.
Scheme 3. Functionalized phenyl acetates that were studied.
=
two small structural changes had to be introduced. The C N
Table 1: Observed pseudo-first-order rate constants for the ester cleav-
age of compounds 3A, 3B, and 3I under different basicconditions.
double bond had to be reduced to render the structure stable
under the basic conditions required for ester cleavage. Such a
covalent post-modification is very common in imine-based
dynamic combinatorial chemistry.[18] Secondly, the resulting
secondary amine had to be methylated to prevent an intra-
molecular attack of the amine on the neighboring ester.
The effect of the presence of the ammonium and urea
groups in 3B and 3I, respectively, on the ester cleavage was
initially studied by measuring the methanolysis rates of 3A,
3B, and 3I because of the similarity to the conditions under
which the amplification studies were performed. Twelve
equivalents of sodium methoxide were added to 0.6 mm
solutions of esters 3A, 3B, and 3I in methanol at room
temperature, and the methanolysis was followed by measur-
ing the increase in absorbance at 280 nm (Figure 2a). The
resulting curves were fitted using a first-order exponential
[a]
Entry[b]
3A
3B (k3B/k3A)
3I (k3I/k3A)
1
2
3
0.82”10À2
1.85”10À5
4.41”10À7
3.96”10À2
(4.8)
(16.4)
(59.9)
1.28”10À2
(1.6)
(1.3)
(11.8)
3.02”10À4
2.64”10À5
2.42”10À5
5.23”10À6
[a] kobs [sÀ1]. Kinetics were followed by UV/Vis spectroscopy at 280 nm.
[b] Entry 1: [3]=0.6 mm, [NaOMe]=7.2 mm, MeOH, 258C; entry 2:
[3]=0.6 mm, pH 11 ([CAPS]=60 mm), H2O/CH3CN=50:50, 458C;
entry 3: [3]=0.6 mm, pH 11 ([CAPS]=60 mm), H2O/CH3CN=10:90,
458C.
ing substrate concentrations yielded the same rate constants,
which is in strong support of intramolecular catalysis. Finally,
if the ester moieties in 3B and 3I are cleaved at a higher rate
owing to a stabilization of the negative charges in the
yielding the pseudo-first-order rate constants given in Table 1, transition state by the neighboring groups (Figure 3), we
entry 1. The resulting rates are in good agreement with the
results of the amplification studies, both in terms of the order
of reactivity (kobs,3B > kobs,3I > kobs,3A) and the relative acceler-
ation (4.8:1.6:1 for 3B, 3I, and 3A, respectively).
should observe an enhanced catalytic effect upon decreasing
the polarity of the medium, because of a
lower solvation ability of the solvent.
Therefore, we performed hydrolysis
studies of compounds 3 both in a 1:1
and a 1:9 mixture of H2O/CH3CN
buffered at pH 11.[19] The observed
order of reactivity for compounds 3 in
the 1:1 mixture is in line with the trend
observed for the methanolysis reactions,
Several control experiments confirm that the increased
ester cleavage rate results from an intramolecular neighbor-
ing-group effect. Firstly, the presence of one equivalent of
tetramethylammonium chloride did not affect the methanol-
ysis rate of compound 3A at all. This excludes the possibility
that the higher rate observed for 3B is simply due to a change
in ionic strength in the mixture. In other words, it shows that
the ammonium ion needs to be present in close proximity to
the carboxylic ester to induce a catalytic effect. Secondly,
measuring the methanolysis rate of compound 3B at decreas-
although the hydrolysis rate for com-
pound 3B is slightly higher than
Figure 3. Stabilization
of the transition state
expected (ca. 16-fold; Table 1, entry 2).
Importantly, however, a decrease of the ysis of 3B.
during the methanol-
Angew. Chem. Int. Ed. 2008, 47, 2475 –2479
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
2477

Communications
added to scaffold 1 or 2 (5 mm) in CD3OD. The mixtures were kept at
508C until the thermodynamic equilibrium was reached (no further
change in the 1H NMR spectra of the mixtures). 1H–13C HSQC
spectra (see Supporting Information) were used to assess library
composition.
b) Couples: Freshly prepared mother solutions of the scaffold
molecule (1 or 2, 100 mm in CD3CN) and the hydrazides (A–I,
500 mm in CD3OD) were used to prepare the mixtures of 1 or
2:A:(B–I) in a 1:5:5 ratio with a final scaffold concentration of 5 mm.
The solutions were kept at 508C and monitored by 1H NMR
spectroscopy until no additional changes were observed in time.
Integration of the hydrazone signals yielded the relative concentra-
tions of the two hydrazones.
General procedure for the kinetic experiments: A stock solution
of 3 (10 mm in CH3CN) was diluted and added to a cuvette containing
either a) a NaOMe (7.2 mm) solution in MeOH at 258C, b) a 1:1, or
c) a 9:1 mixture of CH3CN:H2O containing CAPS (3-(cyclohexyla-
mino)-1-propanesulfonic acid) buffer (60 mm) at 458C, obtaining a
final concentration of 3 of 0.6 mm. Kinetics were followed by
measuring the increase of absorbance at 280 nm in time. Ester
cleavage was confirmed by HPLC and ESI-MS.
polarity of the medium results in a significant jump in
hydrolysis rates both for compounds 3B and 3I relative to the
reference compound 3A. In a 1:9 mixture of H2O/CH3CN
compounds 3B and 3I are now hydrolyzed approximately
60 and 16 times faster, respectively (Figure 2b and Table 1,
entry 3). These results strongly support our hypothesis that
the enhanced reactivity of compounds 3B and 3I is indeed
due to transition-state stabilization.[20,21] The fact that the
ammonium moiety is the best catalytic unit indicates that
there is a considerable amount of charge development in the
transition state and charge–charge interaction prevails over
hydrogen bonding in its stabilization. Accelerations similar to
those that we have found here have been obtained with
imprinted polymers, where catalysis is also based only on
transition state stabilization.[7a,c,22]
Classical studies on intramolecular interactions between
neighboring groups have generally shown the importance of
the geometry of the complex.[23] To assess the influence of the
geometry in this system, we have studied the behavior of
hydrazide B3 and phenyl acetate 3B3,[24] in which the
ammonium group is attached by a propylene rather than a
methylene spacer. A competition experiment between hy-
drazide B3 and reference hydrazide A using both scaffolds 1
and 2 yielded an amplification factor of 1.5, which is slightly
less than that obtained for hydrazide B under the same
conditions (1.8). The same trend is observed for the meth-
anolysis rate of 3B3 (kobs = 3.33 ” 10À2 sÀ1), which is lower than
that obtained for 3B, but still represents an acceleration of 4.1
with respect to reference compound 3A. These results show
that increasing the spacer length reduces the efficiency of the
ammonium group in catalyzing the ester cleavage.[25]
In summary, although the selected catalytic functionality
is hardly unexpected,[21] we have demonstrated the great
potential of the tethering strategy for detecting noncovalent
interactions that play a role in catalysis. Using a phosphonate
target as a transition-state analogue of the hydrolysis of an
ester bound to the reacting aldehyde, we have provided a
proof-of-principle for the self-selection of functional groups
that assist in catalysis. Very simple, commercially available
molecules were used, which also illustrates the scope of this
approach. In terms of developing enzyme-like catalysts, the
tethering strategy, in contrast to currently performed dynamic
screening methods, allows for an independent optimization of
the binding and catalytic events. We have reported herein on
the catalytic unit selection, but research in our laboratory is
currently aimed at selecting also the recognition site to fully
implement the catalyst selection.
Received: August 22, 2007
Revised: November 9, 2007
Published online: February 20, 2008
Keywords: catalysis · combinatorial chemistry ·
.
supramolecular chemistry · thermodynamics · transition states
[1] a) P. T. Corbett, J. Leclaire, L. Vial, K. R. West, J.-L. Wietor,
J. K. M. Sanders, S. Otto, Chem. Rev. 2006, 106, 3652 – 3711; b) J.-
M. Lehn, Chem. Soc. Rev. 2007, 36, 151 – 160.
[2] For theoretical discussions on the amplification in dynamic
libraries: a) J. S. Moore, N. W. Zimmerman, Org. Lett. 2000, 2,
915 – 918; b) Z. Grote, R. Scopelliti, K. Severin, Angew. Chem.
2003, 115, 3951 – 3955; Angew. Chem. Int. Ed. 2003, 42, 3821 –
3825; c) P. T. Corbett, S. Otto, J. K. M. Sanders, Chem. Eur. J.
2004, 10, 3139 – 3143; d) K. Severin, Chem. Eur. J. 2004, 10,
2565 – 2580; e) P. T. Corbett, S. Otto, J. K. M. Sanders, Org. Lett.
2004, 6, 1825 – 1827.
[3] For representative examples see: a) J.-M. Lehn, A. Eliseev,
Science 2001, 291, 2331 – 2332; b) S. Otto, R. L. E. Furlan,
J. K. M. Sanders, Science 2002, 297, 590 – 593; c) R. T. S. Lam,
A. Belenguer, S. L. Roberts, C. Naumann, T. Jarrosson, S. Otto,
J. K. M. Sanders, Science 2005, 308, 667 – 669; d) N. Sreeniva-
sachary, J.-M. Lehn, Proc. Natl. Acad. Sci. USA 2005, 102, 5938 –
5943; e) S. Ladame, A. M. Whitney, S. Balasubramanian, Angew.
Chem. 2005, 117, 5882 – 5885; Angew. Chem. Int. Ed. 2005, 44,
5736 – 5739; f) A. Buryak, K. Severin, Angew. Chem. 2005, 117,
8149 – 8152; Angew. Chem. Int. Ed. 2005, 44, 7935 – 7938.
[4] a) B. Brisig, J. K. M. Sanders, S. Otto, Angew. Chem. 2003, 115,
1308 – 1311; Angew. Chem. Int. Ed. 2003, 42, 1270 – 1273; b) L.
Vial, J. K. M. Sanders, S. Otto, New J. Chem. 2005, 29, 1001 –
1003.
[5] L. Pauling, Chem. Eng. News 1946, 24(161), 707 – 709.
[6] a) D. Hilvert, Annu. Rev. Biochem. 2000, 69, 751 – 793; b) C. V.
Hanson, Y. Nishiyama, S. Paul, Curr. Opin. Biotechnol. 2005, 16,
631 – 636.
[7] a) M. Emgenbroich, G. Wulff, Chem. Eur. J. 2003, 9, 4106 – 4117;
b) J.-Q. Liu, G. Wulff, J. Am. Chem. Soc. 2004, 126, 7452 – 7453;
c) A. Volkmann, O. Brüggeman, React. Funct. Polym. 2006, 66,
1725 – 1733.
Experimental Section
The syntheses and characterization data of compounds B3, 3A, 3B,
3B2, 3B3, and 3I can be found in the Supporting Information,
together with characteristic parts of the H-13C HSQC and H NMR
spectra used for determination of the amplification factors. The
Supporting Information also contains a plot of log kobs as a function of
pH obtained for the hydrolysis of compounds 3A and 3B, the
methanolysis of compound 3B3, and the methanolysis of 3B at
different concentrations.
1
1
[8] A. Fersht in Structure and Mechanism in Protein Science, 3rd ed.,
W. H. Freeman, New York, 2000.
Procedures for the equilibration experiments: a) Libraries: A
mixture of hydrazides A–I (each 7.5 mm final concentration) were
[9] A. J. Kirby, Angew. Chem. 1996, 108, 770 – 790; Angew. Chem.
Int. Ed. Engl. 1996, 35, 706 – 724.
2478 www.angewandte.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 2475 –2479

Angewandte
Chemie
[10] a) D. A. Erlanson, A. C. Braisted, D. R. Raphael, M. Randal,
R. M. Stroud, E. M. Gordon, J. A. Wells, Proc. Natl. Acad. Sci.
USA 2000, 97, 9367 – 9372; b) Y. Krishnan-Ghosh, S. Balasu-
bramanian, Angew. Chem. 2003, 115, 2221 – 2223; Angew. Chem.
Int. Ed. 2003, 42, 2171 – 2173; c) D. A. Erlanson, J. A. Wells,
A. C. Braisted, Annu. Rev. Biophys. Biomol. Struct. 2004, 33,
199 – 223; d) R. M. Bennes, D. Philp, Org. Lett. 2006, 8, 3651 –
3654.
[19] The pH refers to the value of the pure aqueous component, and
was not corrected for the mixture.
[20] In principle, the enhanced hydrolysis rate of 3B compared to 3A
could also be a result of an enhanced local concentration of
OHÀ. If this would have been the case, a plot of logkobs vs pH
would be nonlinear. Therefore, the hydrolysis rates of both 3A
and 3B were determined at pH 7–11 (see Supporting
Information). The analysis showed that for pH 9–11, logkobs
increases linearly with the pH with logk3B and logk3a increasing
linearly with pH with the same slope. Based on these results, we
conclude that the higher hydrolysis rate of 3B is indeed due to
transition-state stabilization. Interestingly, for pH values below
9, the difference in hydrolysis rates between 3A and 3B
vanishes, and the slope of the curve by plotting logkobs versus
pH changes dramatically for 3A. We ascribe this to the
protonation of the tertiary amine in both 3A and 3B which,
for both structures, results in a positive charge located very close
to the carboxylic ester moiety. Consequently, at more acidic pH
values the “catalytic” effect of the ammonium group in 3B is
taken over by the protonated tertiairy amine.
[21] a) R. Fuchs, J. A. Caputo, J. Org. Chem. 1966, 31, 1524 – 1526;
b) J. Hajdu, G. M. Smith, J. Am. Chem. Soc. 1981, 103, 6192 –
6197.
[22] In a structurally related system, Anslyn et al. observed a 40-fold
acceleration of a phosphate diester cleavage by a neighbouring
guanidinium-group. See: A. M. Piatek, M. Gray, E. V. Anslyn, J.
Am. Chem. Soc. 2004, 126, 9878 – 9879.
[23] a) M. I. Page, W. P. Jencks, Proc. Natl. Acad. Sci. USA 1971, 68,
1678 – 1683; b) A. J. Kirby, Adv. Phys. Org. Chem. 1980, 17, 183 –
278; c) G. Illuminati, L. Mandolini, Acc. Chem. Res. 1981, 14,
95 – 102; d) F. Menger, Acc. Chem. Res. 1985, 18, 128 – 134;
e) T. C. Bruice, F. C. Lightstone, Acc. Chem. Res. 1999, 32, 127 –
136. For the importance in a dynamic system see also ref. [10d].
[24] We have also synthesized an analogous compound containing an
ethylene spacer (3B2). Regrettably it is not stable enough to
carry out the hydrolysis studies: addition of base very rapidly
leads to the elimination of trimethylamine.
[11] X. Zhang, K. N. Houk, Acc. Chem. Res. 2005, 38, 379 – 385.
[12] G. Gasparini, M. Martin, L. J. Prins, P. Scrimin, Chem. Commun.
2007, 1340 – 1342.
[13] In principle, the amplification may also be affected by the
occurrence of repulsive or intermolecular interactions.
[14] 1H–13C HSQC spectroscopy can also be used to follow the
kinetics of hydrazone exchange. This allows a full evaluation of
both the kinetic and thermodynamic parameters of a dynamic,
multicomponent library and, in addition, eliminates any problem
in peak assignment. This methodology will be published in due
course.
[15] The lowest concentrations are in the order of 0.25 mm. T he
errors in the amplification factors are estimated to be around
15%.
[16] The amplification in this kind of systems depends on the number
of equivalents of hydrazides added,[12] which might explain the
slightly different amplification factors between the two methods
of screening.
[17] As we have shown earlier,[12] maximum amplification is observed
under very dilute conditions (0.2 mm), as under these conditions,
competing intermolecular interactions are minimal. However, at
such concentrations the exchange kinetics are very slow. There-
fore, for practical reasons we decided to work at an intermediate
5 mm concentration of scaffold. From our previous studies, this
implies a drop in amplification from the maximum value of 3.1 to
the observed intermediate value of 1.8.
[18] a) C. Godoy-Alcꢀntar, A. K. Yatsimirsky, J.-M. Lehn, J. Phys.
Org. Chem. 2005, 18, 979 – 985; b) M. Hochgürtel, H. Kroth, D.
Piecha, M. W. Hofmann, C. Nicolau, S. Krause, O. Schaaf, G.
Sonnenmoser, A. V. Eliseev, Proc. Natl. Acad. Sci. USA 2002, 99,
3382 – 3387; c) S. Zameo, B. Vauzeilles, J.-M. Beau, Angew.
Chem. 2005, 117, 987 – 991; Angew. Chem. Int. Ed. 2005, 44, 965 –
969.
[25] This is in accord with what has been found by studying charge-
charge interactions in flexible systems: Y. Chevalier, P. Perchec,
J. Phys. Chem. 1990, 94, 1768 – 1774.
Angew. Chem. Int. Ed. 2008, 47, 2475 –2479
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 2479